Question

I’m learning about regular expressions for a script I will be writing down the road, but I’ve hit a stopping point. I basically understand what ?= and ?! do, they’re “lookaheads”. To borrow and example: /Win (?=98)/ only matches “Win ” if it is followed by “98” whereas /Win (?!XP)/ would match “Win ” only if it was not followed by “XP” . . . right?

Now I really don’t get the ?: delimiter. And I haven’t found a decent example of it and I’m just really, really confused about it. :/ I understand it’s supposed to match the entire contained pattern or something?

One more thing I’m confused about are backreferences. Here’s on example I found: the regular expression /<(\S+).*>(.*)<\/\1>/ is supposed to match “any tag”. I’m just confused as to what the number “1” refers the browser to . . . is it the first match – in which case I would think it would refer to the < character – or something else?

I’m just now dabbling into the world of regular expressions and would love some clarification on these concepts, thank you all in advance!

Answer 1

Your thoughts on the lookahead assertions are correct.

\1 refers to the first submatch in parentheses, i.e. whatever was matched by the (\S+) in your example. \2 refers to the second (in the example, (.*)) and so on.

?:, on the other hand, means that that set of parentheses should not be tied to a reference like \1. You use it if you need parenthesis for something but don’t really care about getting the matched text later on. So, in the regular expression /(?:abc)def(ghi)/, \1 would not expand to abc (because we switched that off using the ?:), but to ghi.